目前是一名python小白，网上找了一篇强化学习的代码想学学，发现好多这种代码都有这个问题：
Q_observation = self.Q_table[observation, :]
IndexError: only integers, slices (:), ellipsis (...), numpy.newaxis (None) and integer or boolean arrays are valid indices

是代码问题还是我的环境设置问题呀，希望大佬求教，小白不胜感激

def __init__(self, obser_n, act_n, epsilong, gama, alfa):
    self.act_n = act_n
    self.Q_table = np.zeros((obser_n, act_n))
    self.epsilong = epsilong
    self.gama = gama
    self.alfa = alfa

"""
根据当前 Observation选择最优 action,采用 ε-greedy 算法，即：ε 概率探索，否则选择Q值最大的策略
每个observation 用一个数字表示，例如：索引1表示observation（1，1）
"""

def actionChoose(self, observation):
    Q_observation = self.Q_table[observation, :]
    if random.uniform(0, 1) > (1 - self.epsilong):
        return np.random.choice(self.act_n)
    else:
        return self.getMaxOfQtable(observation)

# 　根据当前Observation返回Q值最大的action
def getMaxOfQtable(self, observation):
    Q_observation = self.Q_table[observation, :]
    maxList = np.where(Q_observation == max(Q_observation))[0]
    return np.random.choice(maxList)

#  learn算法，更新Q表格
def learn(self, observation, action, reward, next_observation, next_action, is_done):
    if is_done:
        target_value = reward
        this_value = self.Q_table[observation][action]
        self.Q_table[observation][action] += self.alfa * (target_value - this_value)
    else:
        #   先计算目标值
        target_value = reward + self.gama * max(self.Q_table[next_observation, :])
        #   拿当前值
        this_value = self.Q_table[observation][action]
        #   计算时序差分
        diff = target_value - this_value
        #   更新当前Q
        self.Q_table[observation][action] += self.alfa * diff

def getQtable(self):
    return self.Q_table